Thrust 1.17.0

[alliepiper]

Thrust 1.17.0

Summary

Thrust 1.17.0 is the final minor release of the 1.X series. This release provides GDB pretty-printers for device vectors/references, a new unique_count algorithm, and an easier way to create tagged Thrust iterators. Several documentation fixes are included, which can be found on the new Thrust documentation site at https://nvidia.github.io/thrust. We’ll be migrating existing documentation sources to this new location over the next few months.

New Features

• Add make_tagged_iterator #1586: Add new thrust::make_tagged_iterator convenience function. Thanks to @karthikeyann for this contribution.
• Add unique_count algorithm #1619: Add unique_count algorithm. Thanks to @upsj for this contribution.
• Add GDB pretty-printer for thrust vectors and references #1631: Add GDB pretty-printers for device vectors/references to scripts/gdb-pretty-printers.py. Thanks to @upsj for this contribution.

Bug Fixes

• Fix thrust::reduce_by_key for 2^31 elements #1671: Fixed reduce_by_key when called with 2^31 elements.

Other Enhancements

• Use CUB version of CUB adjacent difference #1512: Use CUB to implement adjacent_difference.
• Use CUB's ScanByKey implementation #1555: Use CUB to implement scan_by_key.
• New Doxybook/Jekyll-based GitHub Pages documentation infrastructure (take two) #1611: Add new doxybook-based Thrust documentation at https://nvidia.github.io/thrust.
• Fix contributing guide link to cmake options #1639: Fixed broken link in documentation. Thanks to @jrhemstad for this contribution.
• Increase contrast of search input text in docs. #1644: Increase contrast of search input text in new doc site. Thanks to @bdice for this contribution.
• Add __forceinline__ to thrust::detail::wrapped_function::operator() #1647: Add __forceinline__ annotations to a functor wrapper. Thanks to @mkuron for this contribution.
• Incorrect example for permutation iterator. #1660: Fixed typo in documentation example for permutation_iterator.
• Add an example that shows how to use custom CUDA streams. #1669: Add a new explicit_cuda_stream.cu example that shows how to use explicit CUDA streams and par/par_nosync execution policies.

---

This discussion was created from the release Thrust 1.17.0.